Integrated design method of aircraft RBF neural network-based fault diagnosis and nonlinear dynamic inversion control

Purpose This paper aims to address the actuator fault problem of the boost-glide rocket based on the proposed adaptive fault-tolerant control method. Design/methodology/approach The lock-in-place, loss of effectiveness and bias of the rocket actuator are equivalently regarded as the generalized drift fault. Based on the radial basis function (RBF) neural network, an adaptive update law is designed to diagnose the generalized drift fault. The nonlinear dynamic inversion control is used to adjust the control command based on the fault diagnosis result, then the control system still has a good performance when the actuator fault occurs. The integrated design method of RBF neural network-based fault diagnosis and nonlinear dynamic inversion control is used to achieve a fault-tolerant control effect. Findings The effectiveness of the proposed adaptive fault-tolerant control method is verified by the numerical simulation and the results show that the proposed method has excellent fault suppression capability. Social implications The proposed method effectively mitigates operational risks and associated economic losses of the boost-glide rocket. Originality/value An equivalent expression for actuator faults and an adaptive fault-tolerant control method based on an RBF neural network are proposed and the integrated design method is used to suppress the fault impact effectively.